In smart environments based on the Internet of Things (IoT), almost all of the object information that is collected by various sensors is time series data, which records the behavior of the objects. Analyzing the correlation between different time series data, other than those in the same time series, is more helpful to discovering their behavioral relations. This has become one of the important current issues in the IoT. To analyze the correlation, a clustering algorithm named the CPCCM (clustering algorithm based on precise correlation coefficient matching) is presented. First, each initial sequence is split into a set of subsequences by adopting a preset sliding window. Then, the correlation coefficients between any pair of subsequence sets from two sequences are resolved. Those pairs that pass some preset Pearson correlation coefficient threshold are clustered. In the CPCCM, a cross-traversal strategy is introduced to improve the search efficiency. The cross-traversal strategy alternatively searches the subsequences in two subsequence sets. To improve the clustering efficiency, in each initial sequence, adjacent subsequences are merged into longer subsequences and replaced by it if they appear in the same subsequence set. Finally, by analyzing practical electric power consumption data, the CPCCM is shown to be promising and able to be applied in similar scenarios. By comparison with the agglomerative hierarchical clustering algorithm, the major contributions of this work is that the clustering quality is improved by using the strategy of precise matching and cross-traversal, and complexity of the algorithm is reduced by merging adjacent subsequences. Therefore, CPCCM can be applied to analyze behavior between different objects in smart environments.

In the Internet of Things (IoT), the online performance of many online services is determined by their distribution resources, which are connected to many different devices. The expected performance of a resource service primarily depends on the optimal use of the service in satisfying end-to-end quality requirements to support its successful execution. Therefore, the performance of a resource service is dynamic and should be discovered as a benchmark to detect a performance anomaly online. A performance anomaly is referred to as a business anomaly because it depends on its usage. The performance is measured by the quality of service (QoS) that is possessed by a resource service. In this paper, an approach based on the resource service QoS is proposed to detect a business anomaly via mining business process data in collaborative tasks in the IoT. First, a resource-service chain (RSC) is considered to be an analysis object because resource services are employed as a “service flow” by a business process. The similarity between any two RSCs is measured according to the QoS indicator values of resource services. Based on the similarity, a clustering algorithm is presented to resolve clustering centers that are considered to be QoS benchmarks. Second, according to the QoS benchmarks of RSCs, the thresholds of QoS indicators of a business anomaly are determined. Third, an algorithm is presented to detect anomalies of the business process. Finally, the proposed approach is illustrated by a simulation experiment. The experimental results show that the approach can be used to effectively detect a business anomaly online.

Though ensuring logical consistency between flows in use case specifications (UCSs) and high‐level business processes is a prerequisite and foundation for generating high‐quality, full‐coverage test cases, this task still primarily relies on manual effort. To address this limitation, a large language model (LLM)‐assisted test‐driven approach is proposed, which introduces a formal structure to guide LLMs in generating UCSs based on natural language requirements. Specifically, to validate the consistency of UCS flows with high‐level business process logic, UML activity and state machine diagrams are used as specific modeling methods for such processes. Furthermore, three novel rules are proposed to validate the consistency: business objects are incorporated to not only horizontally connect UCSs with these models but also vertically bridge different abstraction levels of requirements. To normalize this methodology, a comprehensive framework is established—this framework enables bidirectional traceability between requirements and testing while simultaneously building a feedback loop that integrates requirements, testing, and process models. Experimental results show that the approach not only improves the requirements specification quality but also facilitates semi‐automatic generation of test cases for acceptance testing, thereby enhancing the efficiency of the overall development process.

Accurately grasping electricity consumption behaviour and trend in different period can provide decision support for electric resource scheduling. It is necessary to analyze the electricity consumption time sequence. To serve the purpose, a clustering method called VDSI (Variance of Difference between two Sequence Items) is presented by analyzing the typical daily electricity consumption. In the method, time sequence of electricity consumption is split into sub-sequences based on a predefined time window. Firstly, any two time sub-sequences are converted to a new one, in which any item is the difference between items of the two subsequences. Then, the variance is used to measure the distance of the two sub-sequences. Finally, hierarchical clustering algorithm is used to cluster sequences in different time window. By comparing with traditional sequence clustering, the VDSI can obtain more accurate clustering results.